[Xcb] Invalid XIDs

Nathaniel Smith njs at pobox.com
Sat Feb 21 15:50:13 PST 2009


On Wed, Feb 18, 2009 at 9:13 AM, Barton C Massey <bart at cs.pdx.edu> wrote:
> Sorry to be so late in getting back to you: I got
> sidetracked.  I'm attaching the patch I intend to commit for
> review: I decided I'm not really comfortable with it until
> some others have looked at it, as it's a little involved.

Sure. It looks like there are some other people seeing this problem
much more often than I, who might be able to test it more usefully:
  https://bugzilla.redhat.com/show_bug.cgi?id=458657
  https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/220628

> --- a/src/xcb_xid.c
> +++ b/src/xcb_xid.c
> +            /* XXX The latter disjunct is what the server returns
> +               when it is out of XIDs.  Sweet. */
> +            if(!range || range->start_id == 0 && range->count == 1)
> +            {
> +                pthread_mutex_unlock(&c->xid.lock);
> +                return -1;
> +            }
> +            assert(range->count > 0 && range->start_id > 0);

Is start_id guaranteed to be > 0 here?

-- Nathaniel


More information about the Xcb mailing list